Bear with me, please. Please make sure a highly skilled technical person reads and understands this. I want to describe my vision for (AI/Algorithmically) Optimised Operating Systems. To explain it properly, I will describe the process to build it (pseudo). Required Knowledge (no particular order): Processor Logic Circuits, LLM models, LLM tool usage, Python OO coding, Procedural vs OO, NLP fuzzy matching, benchmarking, canvas/artefacts/dynamic HTML interfaces, concepts of how AI models are vastly compressed and miniaturised forms of full data, Algorithmic vs AI. First, take all OO Python code (example) on GitHub (example), then separate each function from each object into its own procedure (procedural logic) by making a logical procedural list of actions to perform only that function based on its entire dependency chain (i.e. all other objects it relies on). Relate all compiled functions using (for example) fuzzy matching on the name, or AI-based functional profiling to get multiple instances of each function. Starting with the most used function, test each one against the others that perform the same task for bugs and completeness. Determine the fastest, most optimal version of that function (and every function). Add a single instance of each most optimal function to the centralised tool codebase, which will later be utilised by the language models. This ensures we rely only on the most optimised function for each and every use case — with every program using one shared instance of that function instead of compiling it separately. In this way, we optimise all existing (Python OO) code based on maximum availability of certain resources. Use the optimised code to teach the AI how to map each user task directly to optimised code (i.e. each user-facing function in unoptimised program code to its optimised version). Later, teach the codebase how to optimise these functions directly to the logic circuits on the processor — as an Operating System — using essentially the same process. Eventually, use the most optimised form for tasks via a dynamic canvas/artefacts UI interface generated by the Start Menu using the language model. Allow custom interfaces to emerge dynamically from the available functions based on language understanding. Grow and optimise the algorithmic codebase until the entire operating system — with versioning — evolves based on the AI-generated user interfaces (compressing them into new versions of the tools codebase). Please don’t give up until you understand this concept completely and nothing in the description is beyond your full comprehension. It’s a highly significant step forward.

I am excited to share that I have developed a Metal kernel for Flash Attention that eliminates race conditions and fully leverages Apple Silicon’s shared memory and registers. This kernel can dramatically accelerate training of transformer-based models. Early benchmarks suggest that models which previously required months to train could see reductions to just a few hours on Apple hardware, while maintaining numerical stability and accuracy. I plan to make the code publicly available to enable the broader community to benefit. I would be happy to keep you updated on the latest developments and improvements as I continue testing and optimizing the kernel. I believe this work could provide valuable insights for Apple’s machine learning research and products.

Hey Tried using a few regular expressions and all fail with an error: Unhandled error streaming response: A generation guide with an unsupported pattern was used. Is there are a list of supported features? I don't see it in docs, and it takes RegExp. Anything with e.g. [A-Z] fails.

In this WWDC25 session, it is explictely mentioned that apps should support AttributedString for text parameters to their App Intents. However, I have not gotten this to work. Whenever I pass rich text (either generated by the new "Use Model" intent or generated manually for example using "Make Rich Text from Markdown"), my Intent gets an AttributedString with the correct characters, but with all attributes stripped (so in effect just plain text). struct TestIntent: AppIntent { static var title = LocalizedStringResource(stringLiteral: "Test Intent") static var description = IntentDescription("Tests Attributed Strings in Intent Parameters.") @Parameter var text: AttributedString func perform() async throws -> some IntentResult & ReturnsValue<AttributedString> { return .result(value: text) } } Is there anything else I am missing?

When the system language and Siri language are not the same, Apple AI may not be usable. For example, if the system is in English and Siri is in Chinese, it may cause Apple AI to not work. May I ask if there are other reasons why the app still cannot be used internally even after enabling Apple AI?

I am a professional user of the 2019 Mac Pro (7,1) with dual AMD Radeon Pro W6900X MPX modules (32GB VRAM each). This hardware is designed for high-performance compute, but it is currently crippled for modern local LLM/AI workloads under Linux due to Apple's EFI/PCIe routing restrictions. Core Issue: rocminfo reports "No HIP GPUs available" when attempting to use ROCm/amdgpu on Linux Apple's custom EFI firmware blocks full initialization of professional GPU compute assets The dual W6900X GPUs have 64GB combined VRAM and high-bandwidth Infinity Fabric Link, but cannot be fully utilized for local AI inference/training My Specific Request: Apple should provide an official, one-click deployable application that enables full utilization of dual W6900X GPUs for local large language model (LLM) inference and training under Linux. This application must: Fully initialize both W6900X GPUs via HIP/ROCm, establishing valid compute contexts Bypass artificial EFI/PCIe routing restrictions that block access to professional GPU resources Provide a stable, user-friendly one-click deployment experience (similar to NVIDIA's AI Enterprise or AMD's ROCm Hub) Why This Matters: The 2019 Mac Pro is Apple's flagship professional workstation, marketed for compute-intensive workloads. Its high-cost W6900X GPUs should not be locked down for modern AI/LLM use cases. An official one-click deployment solution would demonstrate Apple's commitment to professional AI and unlock significant value for professional users. I look forward to Apple's response and a clear roadmap for enabling this critical capability. #MacPro #Linux #ROCm #LocalLLM #W6900X #CoreML

I am running some experiments with WebGPU using the wgpu crate in rust. I have some Buffers already allocated in the GPU. Is it possible to use those already existing buffers directly as inputs to a predict call in CoreML? I want to prevent gpu to cpu download time as much as possible. Or are there any other ways to do something like this. Is this only possible using the latest Tensor object which came out with Metal 4 ?

Hello, My goal is to enable users to perform a freeform search request for any product I sell using a spoken phrase, for example, "Hey Siri, search GAMING CONSOLES on MyCatalogApp". The result would launch MyCatalogApp and navigate to a search results page displaying gaming consoles. I have defined a SearchIntent (using the .system.search schema) and a Shortcut to accomplish this. However, Siri doesn't seem to be able to correctly parse the spoken phrase, extract the search string, and provide it as the critiria term within SearchIntent. What am I doing wrong? Here is the SearchIntent. Note the print() statement outputs the search string--which in the scenario above would be "GAMING CONSOLES"--but it doesn't work. import AppIntents @available(iOS 17.2, *) @AppIntent(schema: .system.search) struct SearchIntent: ShowInAppSearchResultsIntent { static var searchScopes: [StringSearchScope] = [.general] @Parameter(title: "Criteria") var criteria: StringSearchCriteria static var title: LocalizedStringResource = "Search with MyCatalogApp" @MainActor func perform() async throws -> some IntentResult { let searchString = criteria.term print("**** Search String: \(searchString) ****") // tmp debugging try await MyCatalogSearchHelper.search(for: searchString) // fetch results via my async fetch API return .result() } } Here's the Shortcuts definition: import AppIntents @available(iOS 17.2, *) struct Shortcuts: AppShortcutsProvider { @AppShortcutsBuilder static var appShortcuts: [AppShortcut] { AppShortcut( intent: SearchIntent(), phrases: ["Search for \(\.$criteria) on \(.applicationName)."], shortTitle: "Search", systemImageName: "magnifyingglass" ) } } Thanks for any help!

Hi everyone Im currently developing an object detection model that shall identify up to seven classes in an image. While im usually doing development with basic python and the ultralytics library, i thought i would like to give CreateML a shot. The experience is actually very nice, except for the fact that the model seem not to be using any ANE or GPU (MPS) for accelerated training. On https://developer.apple.com/machine-learning/create-ml/ it states: "On-device training Train models blazingly fast right on your Mac while taking advantage of CPU and GPU." Am I doing something wrong? Im running the training on Apple M1 Pro 16GB MacOS 26.1 (Tahoe) Xcode 26.1 (Build version 17B55) It would be super nice to get some feedback or instructions. Thank you in advance!

While analyzing system power on an M4 Max under GPU-heavy compute workloads, I noticed that the the GPU power reported by powermetrics does not come anywhere close to total system DC power reported by the SMC counter PDTR (as used by utilities like mactop). For example, in a heavy GPU workload, powermetrics would report a 65W idle-load delta on the GPU, but at the same time system DC power would rise by 179W, leaving 114W or nearly 2/3 of total system DC power on a Mac Studio M4 Max unexplained. From measurements, the difference appears to correlate with the amount of on-chip data movement (for example, varying bytes-per-FLOP in the workload changes the observed gap). Using SMC and IOReport, I was able to reverse engineer an energy model for the GPU that explains almost all of the energy flow with less than 2% error on the workload I studied. The result is a simple two-term energy roofline model: P_GPU (GPU_combined term in the plot) ≈ a * bytes + b * FLOPs with: ~5 pJ/byte for SRAM movement ~2.7 pJ/FLOP for compute. Has anyone observed similar behavior, or is there guidance on how GPU power reported by IOReport/powermetrics should be interpreted relative to total system power? In particular, I’m interested in whether certain classes of GPU activity may not be attributed to the GPU component in IOReport. Full details with the methodology and results are available here: https://youtu.be/HKxIGgyeISM

Hello Team, I'm currently working on a proof of concept using Apple's Foundation Model for a RAG-based chat system on my MacBook Pro with the M1 Max chip. Environment details: macOS: 26.0 Beta Xcode: 26.0 beta 2 (17A5241o) Target platform: iPad (as the iPhone simulator does not support Foundation models) While testing, even with very small input prompts to the LLM, I intermittently encounter the following error: InferenceError::inference-Failed::Failed to run inference: Context length of 4096 was exceeded during singleExtend. Has anyone else experienced this issue? Are there known limitations or workarounds for context length handling in this setup? Any insights would be appreciated. Thank you!

After exerting a custom model with nms=True. In Xcode, the outputs show as: confidence: MultiArray (0 × 5) coordinates: MultiArray (0 × 4) I want to set fixed shapes (e.g., 100 × 5, 100 × 4), but Xcode does not allow editing—the shape fields are locked. The model graph shows both outputs come directly from a NonMaximumSuppression layer. Is it possible to set fixed output dimensions for NMS outputs in CoreML?

Here's the result: Very weird.

I have built a MAC-OS machine intelligence application that uses Apple Intelligence. A part of the application is to preprocess text. For longer text content I have implemented chunking to get around the token limit. However the application performance is now limited by the fact that Apple Intelligence is sequential in operation. This has a large impact on the application performance. Is there any approach to operate Apple Intelligence in a parallel mode or even a streaming interface. As Apple Intelligence has Private Cloud Services I was hoping to be able to send multiple chunks in parallel as that would significantly improve performance. Any suggestions would be welcome. This could also be considered a request for a future enhancement.

I am using gemini2.5-flash with SwiftUI. How can I receive a response in JSON?

Hello, I am interested in using jax-metal to train ML models using Apple Silicon. I understand this is experimental. After installing jax-metal according to https://developer.apple.com/metal/jax/, my python code fails with the following error JaxRuntimeError: UNKNOWN: -:0:0: error: unknown attribute code: 22 -:0:0: note: in bytecode version 6 produced by: StableHLO_v1.12.1 My issue is identical to the one reported here https://github.com/jax-ml/jax/issues/26968#issuecomment-2733120325, and is fixed by pinning to jax-metal 0.1.1., jax 0.5.0 and jaxlib 0.5.0. Thank you!

When I use ChatGPT in Xcode, the following error is displayed： It was working fine before, but suddenly it became like this, without changing any configuration. Why?

@Generable enum Breakfast { case waffles case pancakes case bagels case eggs } do { let session = LanguageModelSession() let userInput = "I want something sweet." let prompt = "Pick the ideal breakfast for request: (userInput)" let response = try await session.respond(to: prompt,generating: Breakfast.self) print(response.content) } catch let error { print(error) } i want to test the @Generable demo but get error with below:decodingFailure(FoundationModels.LanguageModelSession.GenerationError.Context(debugDescription: "Failed to convert text into into GeneratedContent\nText: waffles", underlyingErrors: [Swift.DecodingError.dataCorrupted(Swift.DecodingError.Context(codingPath: [], debugDescription: "The given data was not valid JSON.", underlyingError: Optional(Error Domain=NSCocoaErrorDomain Code=3840 "Unexpected character 'w' around line 1, column 1." UserInfo={NSJSONSerializationErrorIndex=0, NSDebugDescription=Unexpected character 'w' around line 1, column 1.})))]))

Is it possible to train an Adaptor for the Foundation Models to produce Generable output? If so what would the response part of the training data need to look like? Presumably, under the hood, the model is outputting JSON (or some other similar structure) that can be decoded to a Generable type. Would the response part of the training data for an Adaptor need to be in that structured format?

The WWDC25: Explore large language models on Apple silicon with MLX video talks about using your own data to fine-tune a large language model. But the video doesn't explain what kind of data can be used. The video just shows the command to use and how to point to the data folder. Can I use PDFs, Word documents, Markdown files to train the model? Are there any code examples on GitHub that demonstrate how to do this?